Current applications of metabolic monitoring in the pediatric intensive care unit

Artificial Neural Network Algorithms to Predict Resting Energy Expenditure in Critically Ill Children

INTRODUCTION

Accurate assessment of resting energy expenditure (REE) can guide optimal nutritional prescription in critically ill children. Indirect calorimetry (IC) is the gold standard for REE measurement, but its use is limited. Alternatively, REE estimates by predictive equations/formulae are often inaccurate. Recently, predicting REE with artificial neural networks (ANN) was found to be accurate in healthy children. We aimed to investigate the role of ANN in predicting REE in critically ill children and to compare the accuracy with common equations/formulae.

STUDY METHODS

We enrolled 257 critically ill children. Nutritional status/vital signs/biochemical values were recorded. We used IC to measure REE. Commonly employed equations/formulae and the VCO₂-based Mehta equation were estimated. ANN analysis to predict REE was conducted, employing the TWIST system.

RESULTS

ANN considered demographic/anthropometric data to model REE. The predictive model was good (accuracy 75.6%; R² = 0.71) but not better than Talbot tables for weight. After adding vital signs/biochemical values, the model became superior to all equations/formulae (accuracy 82.3%, R² = 0.80) and comparable to the Mehta equation. Including IC-measured VCO₂ increased the accuracy to 89.6%, superior to the Mehta equation.

CONCLUSIONS

We described the accuracy of REE prediction using models that include demographic/anthropometric/clinical/metabolic variables. ANN may represent a reliable option for REE estimation, overcoming the inaccuracies of traditional predictive equations/formulae.

Metabolism and energy prescription in critically III children

Optimal nutrition therapy can positively influence clinical outcomes in critically ill children. Accurate assessment of nutritional status, metabolic state, macronutrient requirements and substrate utilization allows accurate prescription of nutrition in this population. In response to stress and injury, the body undergoes adaptive physiologic changes leading to dysregulation of the inflammatory response and hyperactivation of the inflammatory cascade. This results in a global catabolic state with modification in oxygen consumption and macronutrient metabolism. A comprehensive understanding of the metabolic response is essential when prescribing nutritional interventions aimed to offset the burden of this adaptive stress response in the critically ill. In this narrative review we aim to provide a comprehensive review of the physiologic basis, recent literature and some emerging concepts related to energy expenditure and the practical aspects of energy delivery in the critically ill child. Based on the unique metabolic characteristics of the critically ill child, we aim to provide a pragmatic approach to providing nutrition therapy.

Energy Expenditure in Mechanically Ventilated Korean Children: Single-Center Evaluation of a New Estimation Equation

OBJECTIVES

Accurate assessments of energy expenditure are vital for determining optimal nutritional support, especially in critically ill children. We evaluated current methods for energy expenditure prediction, in comparison with indirect calorimetry, and developed a new estimation equation for mechanically ventilated, critically ill Korean children.

DESIGN

Single-center retrospective study.

SETTING

Fourteen-bed pediatric medical ICU in a tertiary care children's hospital.

PATIENTS

Pediatric patients admitted to the PICU between October 2017 and September 2019 with a measured energy expenditure by indirect calorimetry.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

A total 95 pediatric patients (70 in derivation cohort for development of a new predictive equation and 25 in validation cohort) were included. Mean measured energy expenditure of group A was 66.20 ± 15.35 kcal/kg/d. All previously established predictive equations underestimated the predicted energy expenditure, compared with the measured energy expenditure, except the Food and Agriculture/World Health Organization/United Nations University equation. The Schofield-Height and Weight equation showed the best performance among the tested predictive equations for the entire cohort (least bias, -68.58 kcal/d; best percentage, 108.46% ± 33.60%) compared with the measured energy expenditure. It was also the best performing predictive equation in subgroup analysis by age, sex, nutritional status, and organ failure. Because some discrepancies remained between the measured energy expenditure and predicted energy expenditures, we developed a new estimation equation using multiple regression analysis and those variables significantly associated with our current measured energy expenditures: Energy expenditure = -321.264 + 72.152 × (body weight, kg)-1.396 × (body weight) + 5.668 × height (cm) + organ dysfunction* (*hematologic, 76.699; neurologic, -87.984). This new estimation equation showed the least bias and best percentage compared with previous predictive equations (least bias, 15.51 kcal/d; best percentage, 102.30% ± 28.10%).

CONCLUSIONS

There are significant disparities between measured and calculated energy expenditures. We developed a new estimation equation based on measured energy expenditure data that shows better performance in mechanically ventilated Korean children than other equations. This new estimation equation requires further prospective validation in pediatric series with a range in body habitus.

Prealbumin and Retinol Binding Proteins Are Not Usable for Nutrition Follow-Up in Pediatric Intensive Care Units

PURPOSE

Feeding children is a problem in pediatric intensive care units (PICU) and it is difficult to know the correct amount. The purpose of this study is to evaluate if prealbumin or retinol binding proteins (RBP) are effective relative to daily enteral nutrition, without being affected by severity of diseases or infections and can be used to follow up nutritional amount.

METHODS

This is a prospective observational study that includes 81 patients admitted to PICU in Akdeniz University with estimated duration >72 hours, age between 1 month and 8 years. Daily calorie and protein intake were calculated and prealbumin, RBP and C-reactive protein (CRP) levels were measured on the first, third, fifth and seventh mornings.

RESULTS

We find moderate correlation between daily calorie intake and prealbumin levels (r=0.432, p<0.001), RBP levels and daily protein intake (r=0.330, p<0.001). When we investigated the relationship between changes of prealbumin, RBP, CRP, calorie and protein intake during intensive care stay, we found that increase of Prealbumin and RBP levels are explained by decrease of CRP levels (r=-0.546 and -0.645, p<0.001) and not with increase of nourishment.

CONCLUSION

Even adjusted for PRISM3, age and CRP, prealbumin and RBP are correlated with last 24 hours' diet. However, it is not convenient to use as a follow up biomarker because increase of their levels is related with decrease of CRP levels.

The science and art of pediatric critical care nutrition

PURPOSE OF REVIEW

Nutritional status and nutrient delivery during critical illness impact clinical outcomes. We have reviewed recent studies that may guide best practices regarding nutrition therapy in critically ill children.

RECENT FINDINGS

Malnutrition is prevalent in the pediatric ICU population, and is associated with worse outcomes. Nutrition support teams, dedicated dietitians, and educational programs facilitate surveillance for existing malnutrition and nutrition risk, but specific tools for the pediatric ICU population are lacking. Estimation of macronutrient requirements is often inaccurate; novel strategies to accurately determine energy expenditure are being explored. Indirect calorimetry remains the reference method for measuring energy expenditure. Enteral nutrition is the preferred route for nutrition in patients with a functioning gut. Early enteral nutrition and delivery of adequate macronutrients, particularly protein, have been associated with improved clinical outcomes. Delivery of enteral nutrition is often interrupted because of fasting around procedures and perceived intolerance. Objective measures for detection and management of intolerance to nutrient intake are required. In low-risk patients who are able to tolerate enteral nutrition, supplemental parenteral nutrition may be delayed during the first week of critical illness.

SUMMARY

Systematic research and consensus-based practices are expected to promote optimal nutritional practices in critically ill children with the potential to improve clinical outcomes.

Can energy intake alter clinical and hospital outcomes in PICU?

BACKGROUND & AIMS

Energy is essential for the treatment and recovery of children admitted to Pediatric Intensive Care Units (PICU). There are significant immediate and long-term health consequences of both under- and over-feeding in this population. Energy requirements of critically ill children vary depending on age, nutritional status, sepsis, fever, pharmacotherapy, and duration and stage of critical illness. This study aimed to determine the incidence of over- and under-feeding and to compare hospital outcomes between these feeding categories. Secondary outcomes were collected to describe the association between feeding categories and biochemistries (serum lactate, triglycerides, C-reactive protein).

METHODS

An ethics approved retrospective study of children admitted to PICU was performed. All intubated patients admitted to PICU (2008-2013) were included, except those in which an IC test was not feasible. Data collection included demographics, the primary outcome variable reported as under feeding (<90%MREE), appropriate (MREE ±10%) or overfeeding (>110% MREE) determined through comparison of measured resting energy expenditure (MREE) using indirect calorimetry (IC) to actual energy intake based on predicted basal metabolic rate (PBMR) and clinical outcomes mechanical ventilation and PICU length of stay (LOS). Data were analysed with descriptive methods, ANOVA and linear regression models.

RESULTS

A total of 139 patients aged 10 (range 0.03-204) months were included. Sixty (43%) were female and 77 (55%) were admitted after a surgical procedure. A total of 210 IC tests were conducted showing a statistically significant difference between MREE measurements and PBMR (p = 0.019). Of the 210 measurements, only 26 measures (12.4%) demonstrated appropriate feeding, while 72 (34.3) were underfed and 112 (53.3%) were overfed. Children who were overfed had significantly longer PICU LOS (median 45.5, IQR 47.8 days) compared to those children in the appropriately fed (median 21.0, IQR 54.5 days), and underfed groups (median 16.5, IQR 21.3 days). There was a mean difference between the over and under feeding category and ventilation days after adjusting for age and PRISM score (p = 0.026), suggesting decreased mechanical ventilation days for underfed. Children who were underfed had significantly higher CRP (median 75.5, IQR 152.8 mg/L) compared to those children in the appropriately fed (median 57.8, IQR 90.9 mg/L) and overfed groups (median 22.4, IQR 56.2 mg/L).

CONCLUSIONS

This retrospective study confirms that estimations of energy expenditure in critically ill children are inaccurate leading to unintended under and overfeeding. Importantly under feeding seems to be associated with fewer mechanical ventilation days and PICU LOS. Further research is required to elucidate the role of optimal nutrition in altering clinical variables in this population.

Do PICU patients meet technical criteria for performing indirect calorimetry?

BACKGROUND & AIMS

Indirect calorimetry (IC) is considered gold standard for assessing energy needs of critically ill children as predictive equations and clinical status indicators are often unreliable. Accurate assessment of energy requirements in this vulnerable population is essential given the high risk of over or underfeeding and the consequences thereof. The proportion of patients and patient days in pediatric intensive care (PICU) for which energy expenditure (EE) can be measured using IC is currently unknown. In the current study, we aimed to quantify the daily proportion of consecutive PICU patients who met technical criteria to perform indirect calorimetry and describe the technical contraindications when criteria were not met.

METHODS

Prospective, observational, single-centre study conducted in a cardiac and general PICU. All consecutive patients admitted for at least 96 h were included in the study. Variables collected for each patient included age at admission, admission diagnosis, and if technical criteria for indirect calorimetry were met. Technical criteria variables were collected within the same 2 h each morning and include: provision of supplemental oxygen, ventilator settings, endotracheal tube (ETT) leak, diagnosis of chest tube air leak, provision of external gas support (i.e. nitric oxide), and provision of extracorporeal membrane oxygenation (ECMO).

RESULTS

288 patients were included for a total of 3590 patient days between June 2014 and February 2015. The main reasons for admission were: surgery (cardiac and non-cardiac), respiratory distress, trauma, oncology and medicine/other. The median (interquartile range) patient age was 0.7 (0.3-4.6) years. The median length of PICU stay was 7 (5-14) days. Only 34% (95% CI, 32.4-35.5%) of patient days met technical criteria for IC. For patients less than 6 months of age, technical criteria were met on significantly fewer patient days (29%, p < 0.01). Moreover, 27% of patients did not meet technical criteria for IC on any day during their PICU stay. Most frequent reasons for why IC could not be performed included supplemental oxygen, ECMO, and ETT leak.

CONCLUSIONS

In the current study, technical criteria to perform IC in the PICU were not met for 27% of patients and were not met on 66% of patient days. Moreover, criteria were met on only 29% of days for infants 6 months and younger where children 24 months of age and older still only met criteria on 40% of patient days. This data represents a major gap in the feasibility of current recommendations for assessing energy requirements of this population. Future studies are needed to improve methods of predicting and measuring energy requirements in critically ill children who do not meet current criteria for indirect calorimetry.

Controversies in nutritional support for critically ill children

Nutritional support for critically ill infants and children is of paramount importance and can greatly affect the outcome of these patients. The energy requirement of children is unique to their size, gestational age, and physiologic stress, and the treatment algorithms developed in adult intensive care units cannot easily be applied to pediatric patients. This article reviews some of the ongoing controversial topics of fluid, electrolyte, and nutritional support for critically ill pediatric patients focusing on glycemic control and dysnatremia. The use of enteral and parenteral nutrition as well as parenteral nutritional-associated cholestasis will also be discussed.

Energy expenditure: how much does it matter in infant and pediatric chronic disorders?

A sound understanding of energy needs during chronic illness is necessary to avoid imbalances in energy intake and requirements. Failure to accurately estimate energy needs results in both underfeeding and overfeeding in chronically ill children. Suboptimal energy and protein intake may lead to deterioration in body composition, particularly lean body mass loss, which eventually impacts functional outcomes in these vulnerable groups. Furthermore, infants and children with chronic illnesses have a high prevalence of malnutrition and can ill afford further nutritional deterioration from suboptimal nutrient delivery. On the other hand, unintended delivery of excessive energy in patients with chronic respiratory insufficiency results in increased respiratory burden and poor outcomes. Hence, awareness of the energy requirements and attention to energy and protein balance are important when caring for children with chronic illnesses. The basic concepts of the metabolic stress response, measurement of energy expenditure, and the impact of energy imbalance on clinical outcomes in children with chronic illness are reviewed.